Practice free

Compatibility functions

Compatibility functions are older versions kept so old workbooks still calculate. Each one has a newer replacement you should use in new work.

41 functions in this category.

Start here

All compatibility functions

FunctionWhat it doesLevelSinceSpills?
BETADIST Gives the cumulative probability of a beta distribution, used for proportions and project-completion style models bounded between two limits. Expert 2007 or earlier No
BETAINV Reverses BETADIST: given a probability, it finds the point on the beta curve that reaches it. Expert 2007 or earlier No
BINOMDIST Tells you how likely a set number of successes is across repeated yes/no trials, such as defects in a batch or conversions from a mailing. Expert 2007 or earlier No
CEILING Rounds a number up to a chosen multiple, such as the next 0.05 for pricing or the next 15 minutes for billing. Intermediate 2007 or earlier No
CHIDIST Returns the p-value for a chi-square statistic, the upper-tail area beyond x. Expert 2007 or earlier No
CHIINV Finds the critical chi-square value for a chosen significance level. Expert 2007 or earlier No
CHITEST Compares observed counts with expected counts and returns the probability that differences this large are just chance. Expert 2007 or earlier No
CONCATENATE Glues pieces of text and values into one string, for example building a full name or a code from separate cells. Beginner 2007 or earlier No
CONFIDENCE Returns the margin of error to add and subtract from a sample mean, using the normal distribution. Expert 2007 or earlier No
COVAR Measures whether two series move together, averaging the products of their deviations from the mean. Expert 2007 or earlier No
CRITBINOM Finds a threshold count for a binomial process, as used in quality-control acceptance limits. Expert 2007 or earlier No
EXPONDIST Models waiting time between random events such as calls arriving or machine failures. Expert 2007 or earlier No
FDIST Gives the p-value for an F statistic, as in comparing variances or reading ANOVA output. Expert 2007 or earlier No
FINV Finds the critical F value for a chosen significance level. Expert 2007 or earlier No
FLOOR Rounds a number down to a chosen multiple, such as the earlier 5-minute mark or the lower price tier. Intermediate 2007 or earlier No
FTEST Tests whether two samples have similar spread, returning a p-value. Expert 2007 or earlier No
GAMMADIST Models positive, right-skewed quantities such as waiting times or claim sizes. Expert 2007 or earlier No
GAMMAINV Inverse of GAMMADIST, used to find percentile cut-offs. Expert 2007 or earlier No
HYPGEOMDIST Answers questions like the chance of drawing exactly 1 defective unit when sampling 4 from a lot of 20 that holds 8. Expert 2007 or earlier No
LOGINV Finds percentile values of a lognormal variable such as prices or incomes. Expert 2007 or earlier No
LOGNORMDIST Gives the chance a lognormally distributed value falls at or below x. Expert 2007 or earlier No
MODE Finds the value that occurs most often in a data set. Intermediate 2007 or earlier No
NEGBINOMDIST Models the number of misses before you get a set number of hits, such as calls before three sales. Expert 2007 or earlier No
NORMDIST Gives the probability of landing at or below a value on a bell curve, e.g. share of test scores under a cutoff. Intermediate 2007 or earlier No
NORMINV Turns a percentile into a score, e.g. the 95th-percentile demand level. Intermediate 2007 or earlier No
NORMSDIST Converts a z-score into the share of values at or below it. Intermediate 2007 or earlier No
NORMSINV Returns the z-score that cuts off the given lower-tail probability. Intermediate 2007 or earlier No
PERCENTILE Finds the value below which a chosen share of the data falls, such as the 90th percentile salary. Intermediate 2007 or earlier No
PERCENTRANK Shows how far up a data set a value sits, e.g. a score at the 25th percentile. Expert 2007 or earlier No
POISSON Models counts of rare events in a fixed window, like support tickets per hour. Expert 2007 or earlier No
QUARTILE Splits data into four parts for quick spread checks. Intermediate 2007 or earlier No
RANK Ranks a value against a list, such as a salesperson against the team. Intermediate 2007 or earlier No
STDEV Measures how spread out sample data is. Intermediate 2007 or earlier No
STDEVP Spread measure when your data is the entire group, not a sample. Intermediate 2007 or earlier No
TDIST Converts a t statistic into a p-value. Expert 2007 or earlier No
TINV Finds the t cut-off for a confidence interval, e.g. 0.05 for 95%. Expert 2007 or earlier No
TTEST Compares two sets of results, such as before and after, and says how likely the difference is chance. Expert 2007 or earlier No
VAR Squared spread of sample data. Intermediate 2007 or earlier No
VARP Variance when the data covers the whole group. Intermediate 2007 or earlier No
WEIBULL Used in reliability work to model time to failure. Expert 2007 or earlier No
ZTEST Tests whether a sample mean is above a hypothesised value. Expert 2007 or earlier No

Other categories